Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 11:09am 13 Jul 2018
Copy link to clipboard
Print this post
So I already have three home brew inverters, two in service and one as spare, each capable to power my house, why build another one?
Just for fun , its a dreadful malaise that inverter building. Plus I already had many of the key parts on hand.
This one is for experimenting, designed to be able to change parts very easily and those of the blow up variety just requiring unscrewing. No more messing up the PCB with un- soldering...
To help those a bit new to this inverter home brewing I'll start with the mechanical side of the heat sink preparation. If this is nothing new to you just skip over it, I hope its of use to some builders.
My heatsink came from a burnt out 1KW Orion GTI, water ingress had converted the inside into a charred mess. But the heatsink looked usable so I extracted it. It looked like this after a clean up:
Note the pitting where the sparks did their mischief.
The black anodising has to come off anyway so I used a random orbital sander, going up from 80 grid pads to 240 grid which looked fine enough.
The 'pits' turned out not too bad and were just at one area.
The heatsink was then cut to size, first right at the middle fin trough, giving me two heatsinks exactly the same size. I used a table top circular saw with an aluminium blade (important!) for this job and also lanolin spray for cutting lubricant. A full face shield is a must for this job.
Note the extended base at the outsides, this is important with this build, more about that later. Next, one half section was again cut in half.
I now had the three heatsinks ready for more work. The longer heat sink is 11.5" long and the fins are 30mm long. I designed my PCB with this in mind.
As the extended base edge becomes a contact surface against my PCB it had to be stripped of its anodising. I used a spacer to clamp the boards parallel some distance apart so the random orbital sander had a wide base to cut squarely.
Then came the marking out of the required tapped holes. I use a clutch pencil for this, it makes a fine line for the high accuracy required for this build. After pencil marking all the holes I found it best to put the heatsinks aside for a day or two, just in case I change my mind . You might spot on a subsequent picture the rubbing out marks from the eraser, this is sooo much easier than removing scriber grooves or, worse, center punch marks.
With the holes marked at their final position I used a sharp scriber point to precisely indent the heatsink where the lines cross, using just hand pressure. Then a spring loaded automatic center punch was employed to mark the hole. A manual punch works just as well if due care is taken (use a smallish hammer).
Now the drilling process begins. This is a job for a bench or pedestal drill - forget using handheld drills, the holes will never line up using that.
The heatsink is clamped into a suitable drill vice and I begin with 2mm pilot holes. The holes on the flat side of the heatsink were aligned to fall into a 'valley' between the fins, making them through holes and easier to tap later. The holes at the extended base edge are blind holes, drilled 15mm deep.
I used Trefolex spray lubricant, backing the 2mm drill out often to clear the swarf from the flutes. Next came the drill bit to suit the tap, I used 2.5mm for M3 and 3.3mm for M4. Afterwards the top of the holes were slightly countersunk to stop a burr forming there during tapping.
I use this gadget to align the tap square to the work, made that many years ago from a aluminium door stop which was just the right size. Varying inserts were made for this, to accommodate the different tap shanks.
The tap wrench shown can be turned between thumb and fore finger right at the shaft for minimum torque, very important for smallish taps like M3.
A broken tap will ruin the whole day as well as the heat sink so a lot of care was taken. This is not a job to hurry up.
I turn the tap 1/2 turn then back out 1/4 turn to break the swarf, repeat. After 4-5 full turns the tap is carefully backed out completely and the swarf rinsed off in a little jar full of kerosine. This also lubricates the tap.
The tap often 'jams' when trying to back out, the swarf stuck in there is doing that, especially with blind holes. Patience and careful turning back and forth just a little bit will free it so it can be backed out fully. Taps are very brittle and break easily .
To be continued.
Klaus
johnmc Senior Member Joined: 21/01/2011 Location: AustraliaPosts: 282
Posted: 12:31am 14 Jul 2018
Copy link to clipboard
Print this post
Watching your new endeavour Tinker.
Cheers johnjohnmc
brucedownunder2 Guru Joined: 14/09/2005 Location: AustraliaPosts: 1548
Posted: 01:25am 14 Jul 2018
Copy link to clipboard
Print this post
Very Informative ,Klaus..
You must have done a technical training course way back when young guys actually made things,lol.???
Nice work,
BruceBushboy
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 01:33am 14 Jul 2018
Copy link to clipboard
Print this post
Definitely agree that after marking out, its often the case of thinking about how things could be improved or simplified in some slight way. My own method is to first lay down masking tape and then use a very fine line black pen. If I change my mind, just peel off the masking tape and start afresh.Cheers, Tony.
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 08:51am 14 Jul 2018
Copy link to clipboard
Print this post
Thanks Bruce, I did a 'proper' apprenticeship, started way back in 1960. In them days we were taught to be proud of our work and do the best we could, none of that "she'll be right" stuff kids grow up with these days.
Good idea Tony that masking tape trick. Unfortunately at my place fine black pens seem to have run dry whenever I want to use one. Clutch pencils last longer . Klaus
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 09:58am 14 Jul 2018
Copy link to clipboard
Print this post
I'll continue this tale with a short mention of the transformer I used. Lots has been written here how to convert one and I have three full conversions under my belt.
For this one, donated from a defunct Latronics 96V 2.5KVA LF Inverter, I got lazy and only carefully unwound the 2mm diameter primary winding. This was then straightened and wound back on as a 4 in hand winding with a total wire area of 12.5mm sq. Possibly good for 50 Amps. The 230V secondary, having 1.8mm wire (2.5mm sq), was left in place. So was a feedback winding producing a suitable low voltage - more about that later.
Thanks to tinyt for that feedback idea, it makes a lot of sense to me getting the feedback from the very transformer rather than second hand from another transformer.
Those just starting such a project might consider including a feedback winding, its not at all critical since only about 6 Volt are required and that translates to less than ten turns, using any spare thin enameled wire, for most toroid or other transformer cores.
Here is what my toroid now looks like:
Another part that's required are those big capacitors. I used the screw terminal type on my last two inverters, They are sooo much easier to fit or remove as the case may be. However, they are horribly expensive, even the second hand ones offered on Ebay. And one requires 4 or more of these.
So I decided to make my own by soldering brass hex threaded spacers to the common solder terminal type capacitors. My PCB was laid out on a 0.1" grid and as the hex spacers must line up precisely with the PCB holes for the screw terminals I required a jig for this.
I made this from a narrow strip of PCB fiberglass base, this withstands the soldering heat, it does not suck away heat as metal would do and can be precisely drilled for the two 4mm holes to be 0.4" apart.
The spacers are positioned so they are centered over and hard against the 'rivet' heads of the original solder terminals.
It is *very* important to completely discharge the capacitors before doing anything to its terminals. Otherwise a nasty surprise awaits. You can guess how I know that .
The capacitor and the jigged hex spacers have to be kept firmly in place during the soldering process. The soldering has to be quick, I did not want to 'slow cook' the innards of the capacitors. I used rosin paste to coat the solder area and a 60W iron with a chisel tip to do the soldering. Its a tricky job, not much access and nothing must move as the solder cools down (changes from shiny to dull).
Since the original solder terminals were not particular beefy, care is required when unscrewing the jig. I used long nose pliers to hold the hex spacer while unscrewing.
Now for the next step. First masking tape was wrapped around the capacitor base so it protruded about 2 mm.
Then the M4 x 12mm screws were inserted again, I had used M4 x 12mm tapped spacers. This is to prevent the epoxy seeping up inside the thread and clogging it up - it will, believe me . To stop the screw threads from also being epoxied in they are first dunked in a jar of Vaseline before inserting them.
Then the 'moat' is filled with epoxy up to the masking tape rim. This needs to harden *very* well, best to leave it twice as long as the recommended time.
Lastly, using the long nose pliers again, the screws are removed and, hey presto, there is a set of screw terminal capacitors.
To be continued.
Edited by Tinker 2018-07-15Klaus
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 10:48am 14 Jul 2018
Copy link to clipboard
Print this post
Klaus, that is pure genius !!Cheers, Tony.
renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678
Posted: 12:07pm 14 Jul 2018
Copy link to clipboard
Print this post
Very helpful tips you have provided so far Klaus. Just curious, why the desire to have screw in caps?Cheers Caveman Mark Off grid eastern Melb
tinyt Guru Joined: 12/11/2017 Location: United StatesPosts: 438
Posted: 01:59pm 14 Jul 2018
Copy link to clipboard
Print this post
Watching the work of a genius with great interest, very informative. That tapping gadget is very clever, maybe you can provide more info.
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156
Posted: 02:04pm 14 Jul 2018
Copy link to clipboard
Print this post
Watching with interest Klaus - Thanks for taking the trouble to share the build with us. What are the dimensions of your toroid with windings ? I have a Toroidal transformers possibly from a solar feed inverter. It is about 250mm diameter and around 85mm high. It has one part number on it so I will post the picture, maybe someone will recognise it. Not sure if secondary was intended for 230V or 240V either....
Can anyone point me to a good method to work out/establish a reasonable/nominal power through-put for a toroid like this and what sort of % increase over nominal could I pull for say a 10 a minutes overload. My gut feel is it should be good for at least 1.5 - 2kW but also not sure how the HF drive affects a toroid. I hope to leave the secondary intact and just rewind a primary to suit a 48V input ( ~30V ? )
What is the reason that most designs use a separate transformer for voltage feedback from the secondary (240V) output to the EG8010 ? Instinct tells me a small low power overwind from the toroid could provide the same signal (as good or better?) with less loss & 1 less part to worry about ?
Edited by wiseguy 2018-07-16If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 04:20pm 14 Jul 2018
Copy link to clipboard
Print this post
Wiseguy, Power output is only limited by (safe) temperature rise. And that is more a matter of judgment than anything else, but a pretty safe rule of thumb might be four amps per mm squared for the wire size, multiplied by the voltage of the winding.
What sort of overload, and for how long, is again tied to temperature rise. If it feels or smells really hot, it probably is too hot.
The output voltage monitoring needs an isolated low voltage winding for safety reasons, a small extra transformer works, but the extra winding on the toroid is an excellent idea.
The toroid should never see very much high frequency PWM, its the function of the series choke to average out the PWM into a nice smooth 50Hz sine wave before it gets to the primary. If you decide to not use a choke, everything will run slightly hotter, and less efficiently and at higher stress levels.Cheers, Tony.
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 09:34pm 14 Jul 2018
Copy link to clipboard
Print this post
If that toriod is similar to the ones many of us are using they were designed to operate at 3KW continuos in sealed box with no cooling. With forced fresh air to cool it in my experience, you can run closer to 4KW continuous, for periods of about 3 minutes about double that again. Startup surges could be well above 10KW, but all this is assuming it is similar to the 3KW AeroSharp Toroids which have the same external measurements. The EG8010 IC has provision for a temperature sensor, if you use it with the sensor placed on the toroid it will shut down the inverter at about 80 degrees Celcius. Keeping the heatsinks cool is far easier, the toroid is far harder to fix if there was an over temperature failure than if the electronics failed.
If a separate winding was used to provide voltage feedback would it be as responsive to voltage changes as a transformer running off the primary?
There are only 10 types of people in the world: those who understand binary, and those who don't.
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 09:40pm 14 Jul 2018
Copy link to clipboard
Print this post
[quote]If a separate winding was used to provide voltage feedback would it be as responsive to voltage changes as a transformer running off the primary? [/quote] Yes.Cheers, Tony.
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 09:42pm 14 Jul 2018
Copy link to clipboard
Print this post
That would be for fast disassembly when the smoke escapes! along with screw in MOSFETs Edited by Madness 2018-07-16There are only 10 types of people in the world: those who understand binary, and those who don't.
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 09:54pm 14 Jul 2018
Copy link to clipboard
Print this post
Suppose you had a spare circuit board all ready to go, except it was missing some of the larger parts like heatsinks, mosfets, and the large electrolytics.
If everything bolts or screws together, you could probably get that inverter working again in about fifteen minutes. And if you had no power at all, a soldering iron may not be an option anyway.
Cheers, Tony.
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 10:02pm 14 Jul 2018
Copy link to clipboard
Print this post
Circuit boards don't cost that much so having a fully assembled and tested set of PCBs ready to go is an even better option. Since I have done this the spare set has never been needed.There are only 10 types of people in the world: those who understand binary, and those who don't.
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 10:17pm 14 Jul 2018
Copy link to clipboard
Print this post
One of Murphy's laws.Cheers, Tony.
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 08:36am 15 Jul 2018
Copy link to clipboard
Print this post
I will take some pictures of it tomorrow.Klaus
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904
Posted: 08:42am 15 Jul 2018
Copy link to clipboard
Print this post
Agreed, a spare board is the way to go if you want to fix a faulty inverter.
But, as I mentioned at the beginning of this thread, this inverter is for the experimenter. And I like to be able to replace parts or try different ones without wrecking the board or having to grab a new PCB each time. Klaus
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498
Posted: 08:51am 15 Jul 2018
Copy link to clipboard
Print this post
Sort of like a high current breadboard?There are only 10 types of people in the world: those who understand binary, and those who don't.
Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
Print this post
, its a dreadful malaise that inverter building. Plus I already had many of the key parts on hand.
.
.
